Recent trials of gene therapy have focused on the complementation of genetic defects with exogenous healthy copies of mutant genes. While this strategy has shown promise in curing genetic deficiencies, the incorporation of exogenous genes into the genome is associated with a high risk of insertional mutagenesis and oncogenesis. We propose an alternative gene therapy strategy that specifically corrects disease causing mutations. This strategy uses designed variants of existing highly sequence-specific homing endonuclease proteins to cleave mutant genes to promote recombinational repair. Preliminary studies have shown that the DNA recognition specificity of homing endonuclease proteins can be rationally designed towards a chosen target using structure-based computational methods. We have developed tools that use existing homing endonuclease specificity data to locate sequences within genes of interest that are most amenable to cleavage. We will focus on pyruvate kinase deficiency (PKD) as a model disease. PKD is the most common human heritable glycolytic defect, and is the second most common cause of hemolytic anemia. It represents an ideal target disease for repair for several important reasons: the mutations in the pyruvate kinase gene (PKLR) that cause PKD are known;cells to investigate repair are available;and these cells are from an established PKD canine model, which would allow for future elaboration of our work in a large animal transplantation model that closely mimics human physiology. Furthermore, PKD is similar to most genetic diseases in that successfully repaired cells will not have a cellular growth advantage. Since the molecular and cellular character statistics PK deficiency are similar to the majority of common human genetic defects, what we learn about targeted repair of PKD should be broadly applicable to the targeted repair of a wide range of other human genetic diseases. The aims of our project are to design homing endonuclease variants to cleave targets within the PK gene, and to demonstrate that these variants can be used to promote recombinational repair of PK mutations in mammalian cells. The development of homing endonucleases as an adaptable sequence-specific technology has the potential to enable the targeted repair of the majority of human genetic defects. We hope that our studies will serve as a starting point for the development of novel gene therapies that will improve the treatment of a wide range of human genetic diseases.